Approximately 60% of all squamous cell skin cancers appear to arise from pre-existing actinic keratoses (AK) or contiguous skin surfaces. It has been estimated that up to 75% of the white population will develop AK before the age of 86. The exact rate of malignant transformation of AK to squamous cell cancers has not been established, but some investigators have suggested transformation rates of 6-10% over a period of ten years and we have observed a significant 3% per year incidence of squamous cell cancers and a 6% per year incidence of basal cell cancers in AK patients in our recently completed AK study (Thomas Moon, personal communication). There is considerable controversy concerning both the type of AK that require treatment as well as the most efficient treatment program. Cryosurgical application of liquid nitrogen remains the most common and conservative treatment approach, but it is not an innocuous procedure. Although topical fluorouracil is a FDA approved short-term treatment of AK, long-term management with this agent is associated with a high incidence of dose-limiting toxicity and decreasing compliance rates. Clearly, less toxic and more cost-effective chemopreventive strategies are needed in patients with AK as well as multiple squamous and basal cell ski n cancers. We have developed three novel strategies to test scientific hypotheses in normal subjects at high risk for skin cancer and subjects with AK and multiple resected squamous and basal cell skin cancers: 1) Synthetic analogues of alpha-MSH are capable of inhibiting UVB-induced skin damage and converting pheomelanin to eumelanin in normal subjects without significant short or long-term toxicities; 2) topically administered chemopreventive compounds, such as DFMO, vitamin E (free alpha-tocopherol), and/or epigallocatechin gallate (green tea extract) are associated with little or no local or systemic toxicity in human subjects and can reduce BrdU and/or PCNA epidermal cell labeling indices and possibly reverse established AK; and 3) retinyl palmitate in comparison with retinol can be administered in relatively high doses safely with no or minimal side effects and can reduce BrdU and/or PCNA labeling indices and can possibly reverse established AK in subjects with resected squamous and/or basal cell skin cancers. The results of these phase II cancer prevention studies will lead ultimately to phase III trials of the most promising intervention strategies in high risk (for UV-B skin carcinogenesis) subjects with AK and/or resected squamous and basal cell skin cancers.